ventilatory response during dissociative sedation in children—a pilot study

140 Kim et al. d VENTILATORY RESPONSE DURING DISSOCIATIVE SEDATION

Regular article

Ventilatory Response during Dissociative Sedationin Children—A Pilot Study

Grace Kim, MD, Steven M. Green, MD, T. Kent Denmark, MD,Baruch Krauss, MD, EdM

AbstractObjectives: It is unclear whether ketamine induces sub-clinical respiratory depression when administered in disso-ciative doses intravenously (IV). The authors report a pilotstudy of capnography in emergency department (ED)pediatric patients receiving ketamine alone for proceduralsedation, and describe serial measures of ventilatory re-sponse [end-tidal carbon dioxide (EtCO2), respiratory rate,pulse oximetry]. Methods: The authors performed contin-uous capnography on a convenience sample of 20 EDpediatric patients who received ketamine 1.5 mg/kg IV forprocedural sedation. Results: Continuous EtCO2 and pulse

oximetry remained essentially unchanged following keta-mine injection, and no EtCO2 levels . 47 mm Hg werenoted at any point throughout sedation. Conclusions: Nohypoventilation was observed in 20 ED pediatric patientsreceiving ketamine 1.5 mg/kg administered IV over 1minute. The authors found no evidence of respiratorydepressant properties for this dissociative agent. Keywords: ketamine; dissociative sedation; sedation; respira-tory depression; ventilatory response; procedural sedation;capnography. ACADEMIC EMERGENCY MEDICINE2003; 10:140–145.

Ketamine has been established as a safe and effectiveagent for procedural sedation and analgesia (PSA)in the emergency department (ED).1–5 Ketamine istechnically easy to administer by the intramuscular(IM) or intravenous (IV) routes, maintains upperairway tone and protective airway reflexes, andreliably produces potent analgesia, sedation, andamnesia.1–5

A hallmark of the ketamine dissociative state isreliable preservation of baseline respiratory param-eters, including respiratory rate, tidal volume, andminute ventilation.6–13 However, occasional reportsof transient respiratory depression or apnea havebeen described immediately following IV3,13–15 orIM2,16–18 administration; all events were brief andwithout adverse outcome. These rare episodes ofrespiratory depression are believed to result fromunusually high central nervous system concentrationsof ketamine, due to either bolus IV administrationor uncommonly rapid absorption.2,3 However, theysuggest the possibililty that subclinical respiratorydepression may be frequent.

The specific ventilatory response to ketamine inchildren remains poorly defined, including the fre-quency of subclinical hypoventilation. Knowledge ofthe prevalence and pattern of such responses mightassist in predicting, identifying, and avoiding rareoccurrences of clinically important respiratory de-pression during PSA. Two small studies noted sta-bility of end-tidal carbon dioxide (EtCO2) and otherrespiratory parameters during ketamine adminis-tration6,7; however, neither study is representative ofED ketamine administration in that the first usedtight-fitting face masks6 and the second used co-administered midazolam and low (0.5 mg/kg) in-termittent doses of ketamine.7 Also, neither study hadaccess to the newer-generation, sidestream capnog-raphy technology that has been shown to be highlyaccurate in children and does not underestimateEtCO2 as did earlier models.19,20

In order to describe the ventilatory response ofchildren receiving ketamine in standard ED dissocia-tive doses2 and to determine whether hypoventilationof potential clinical importance occurred in thissetting, we continuously measured specific ventila-tory parameters (EtCO2, respiratory rate, pulse oxi-metry) in a pilot sample of children undergoing PSA.

METHODS

Study Design. This study was a prospective obser-vational case series, and was approved by the hospitalinstitutional review board.

Study Setting and Population. The study wasperformed in the ED of an integrated university

From the Department of Emergency Medicine, Loma LindaUniversity Medical Center and Children’s Hospital, Loma Linda,CA (GK, SMG, TKD); and the Division of Emergency Medicine,Children’s Hospital and Harvard Medical School, Boston, MA (BK).Received June 4, 2002; revision received September 20, 2002;accepted September 20, 2002.Oridion Medical loaned a capnograph for this study, but providedno direct financial support.Address for Grace Kim, MD, Loma Linda University MedicalCenter A-108, 11234 Anderson Street, Loma Linda, CA 92354.Fax: 909-558-0121; e-mail: [email protected].

medical center and children’s hospital. We includeda convenience sample of children aged 12 monthsthrough 15 years selected for IV ketamine PSA byeither of two study authors in accordance with theirstandard practice. We excluded children with stan-dard ketamine contraindications1–5 and those who

were American Society of Anesthesiologists PhysicalStatus 3 or greater.

Study Protocol. Ketamine was administered usinga loading dose of 1.5 mg/kg administered overexactly 60 seconds1–5 measured with the second handof a watch. All children were pretreated with IVatropine (0.01 mg/kg, minimum 0.1 mg, maximum0.5 mg), and none were given midazolam or othersedatives.21 Additional doses of ketamine could begiven later at the discretion of the treating physicianas needed to either enhance or prolong sedation. Eachchild had a nurse dedicated to interactive monitoring.Supplemental oxygen was not routinely administered,and applied only if hypoxemia was noted.

Measurements. We measured respiratory parametersusing a low-flow sidestream dual-parameter pulseoximeter and capnograph with Microstream technol-ogy (Oridion Medical, Jerusalem, Israel) that dis-played continuous readings of respiratory rate, heartrate, EtCO2, oxygen saturation, and the CO2 wave-form.19 We attempted capnography prior to ketamineadministration to measure baseline values, andcontinued it throughout sedation until recovery waswell established. We excluded subjects in whom poor

Figure 1. Capnography difficulty (n ¼ 20). *Investigator ratings were made on unmarked 100-mm visual analog scales with the 0 mmlabeled as ‘‘No difficulty’’ and 100 mm labeled as ‘‘Worst possible difficulty.’’

TABLE 1. Characteristics of the Study Children(n 5 20)

Age in years—median (range) 6.5 (1.7–13.0)

Male gender (%) 8 (40%)

ASA* physical status classificationClass 1 20Class 2 or greater 0

Procedural indicationFracture reduction 12Laceration 5Abscess incision and drainage 1Central venous access removal 1Wound exploration 1

Mouth breathing0 attempts needed to correct 81 attempt needed to correct 32 attempts needed to correct 33 or more attempts needed to correct 6

*ASA ¼ American Society of Anesthesiologists.

ACAD EMERG MED d February 2003, Vol. 10, No. 2 d www.aemj.org 141

cooperation precluded reliable baseline measure-ments.

Mouth breathing impairs the accuracy of readingsthrough sidestream systems using nasal cannulae forCO2 sampling, and the Oridion monitor is equippedwith an alarm to indicate when insufficient CO2 isdetected. When such alarms were noted, or when sub-stantial mouth breathing was observed, we attemptedto gently close the child’s mouth and reinitiate apattern of nasal breathing. Oral–nasal cannulae com-patible with the low-flow Microstream system werenot available at the time of the study.

The capnograph was linked to a laptop computerand continuous time-referenced measurements ofCO2 and oxygen saturation (40 measures per second),with concomitant ongoing calculation of EtCO2,respiratory rate, and heart rate, were downloadedinto an electronic database. Treating physicians usedspecified computer keys to mark the database as towhen ketamine administration began and ended.Physicians simultaneously recorded times and eventnotations onto a dataform for corroboration andexplanation. Given our expectation that any respira-tory variation would occur early in the sedationcourse,2–7 we discontinued electronic recording ofcardiopulmonary parameters after 20 minutes, orearlier when initiation of recovery was clinically

apparent. Following discontinuation of electronicrecording, the same cardiopulmonary parameterscontinued to be monitored with visual displays.

Treating physicians completed dataforms concur-rent with patient care that included demographicinformation, procedural indication, American Societyof Anesthesiologists (ASA) physical status,1 and thepresence or absence of adverse events. We definedclinically important airway complications as thoseassociated with a decrease in oxygen saturation below90% and/or those associated with a specific interven-tion (e.g., head repositioning, stimulation, oxygen,assisted ventilation).

Physicians also assessed whether the child experi-enced difficulty tolerating the cannulae pre-sedation,difficulty tolerating the cannulae during sedation,difficulty with mouth breathing, and difficulty main-taining continuous capnography. These ratings weremade using a 100-mm unmarked visual analog scalemarked ‘‘No difficulty’’ at one end and ‘‘Worst pos-sible difficulty’’ at the other. The number of attemptsto correct mouth breathing was recorded.

We defined hypoventilation of potential clinicalimportance as an EtCO2 $ 50 mm Hg.

Data Analysis. We used descriptive statistics toanalyze the data (Stata 7, Stata Corporation, College

Figure 2. Serial end-tidal carbon dioxide (EtCO2) levels. Each dot represents an individual child’s value displayed here at 15-secondintervals. Many points overlap each other. Smoothing line generated by LOWESS (locally weighted least squares) locally weightedregression with a bandwidth of 10.


Station, TX). Given the unavailability of baseline dataon ventilatory response with ketamine, we wereunable to perform a sample size calculation. There-fore, we selected 20 subjects for this pilot study.

RESULTS

The two study physicians attempted data collectionon 27 children over a four-month study period. Onesubject was excluded because poor cooperation pre-cluded baseline capnography, and six others wereexcluded due to technical problems with the com-puter interface that rendered their data unusable.General information for the remaining 20 children isshown in Table 1. Distributions of age and proceduralindication were similar between included and ex-cluded children, and there were no adverse events inthose excluded.

Children generally tolerated the nasal cannulae(Figure 1); however, 12 of 20 children requiredinterventions for mouth breathing (Table 1).

The single airway complication occurred in a 11-year-old boy who developed partial airway obstruc-tion shortly after ketamine administration associatedwith hypoxemia (lowest oxygen saturation 87%).Normal ventilation was promptly restored with

airway repositioning, and assisted ventilation wasnot required. His highest EtCO2 was 41 mm Hg. Theonly other adverse event occurred in a 26-month-oldgirl who experienced a 30-second period of brady-cardia (lowest heart rate 48 beats/min) near theconclusion of her ketamine injection. There was noassociated hypoventilation (highest EtCO2 20 mm Hg)or hypoxemia (lowest 97%), and the pulse quicklyreturned to its baseline between 140 and 150 beats/min without intervention. The precipitant for thebradycardic episode was not determined. No otheradverse events were detected.

Serial EtCO2 and pulse oximetry measures re-mained essentially unchanged following ketamineinjection, excepting resolution of hyperventilation inthe few children with low EtCO2 presedation (Figures2 and 3). Respiratory rates were mildly increased(Figure 4). At no point did any children in the studymeet our definition of hypoventilation of potentialclinical importance (EtCO2 $ 50 mm Hg).

DISCUSSION

To the best of our knowledge, this is the first study ofcapnography in ED children receiving ketamine alonefor PSA. We did not observe subclinical respiratory

Figure 3. Serial pulse oximetry values. Each dot represents an individual child’s value displayed here at 15-second intervals. Manypoints overlap each other. Smoothing line generated by LOWESS (locally weighted least squares) locally weighted regression witha bandwidth of 10. The five data points #90% all occurred in the 11-year-old boy with hypoxemia described in the text.


depression (i.e., hypoventilation) at any point duringsedation of these children. Our findings are consistentwith previous studies documenting reliable preserva-tion of baseline respiratory parameters with ket-amine.6–13 Those prior studies using capnographyalso failed to demonstrate any pattern of hypoventi-lation, although these were in non-ED settings withdifferent administration protocols.6–8

Our study was prompted by rare reports oftransient respiratory depression or apnea immedi-ately following IV3,13–15 or IM2,16–18 administration.Although all such cases have been brief and withoutadverse outcome, they raise reasonable speculationthat subclinical respiratory depression might becommon with ketamine. Our data suggest that thisconcern is unfounded, and that ketamine is nota respiratory depressant even when administered infully dissociative doses (1.5 mg/kg IV over 1 minute).The basis for these rare episodes of respiratory de-pression remains unclear, but may result from un-usually high central nervous system concentrationsof ketamine, due to either bolus IV administration oruncommonly rapid absorption.2,3 In accordance withstandard recommendations, we administered IVdoses over a full minute.1,3,5,6

We are aware of no published evidence to supportan aged-based differential ventilatory response to

dissociative sedation in children. Our data areconsistent with this, in that we found no hypoventi-lation at all ages studied.

The nasal cannulae required for capnography weregenerally well tolerated by our patients, althoughdifficulties with mouth breathing were a minornuisance. In no case did they preclude capnography,however.

Capnography for PSA is a relatively new monitor-ing modality in emergency medicine. Miner andcolleagues found that capnography identified respi-ratory depression more quickly than other monitoringtechniques in 74 adults receiving non-dissociativePSA.22 McQuillen and Steele noted a mean EtCO2

increase of 7 mm Hg in 106 children receiving variousPSA regimens; however, they included no childrenreceiving ketamine without benzodiazepines.23

The single occurrence of transient bradycardia inour study remains unexplained, as hypoxemia wasabsent and no basis for vagal stimulation was noted.Bradycardia is not a complication previously associ-ated with ketamine.1–5

LIMITATIONS

This pilot study is limited by its small sample size.Although rare ventilatory responses are possible, we

Figure 4. Serial respiratory rates. Each dot represents an individual child’s value displayed here at 15-second intervals. Many pointsoverlap each other. Smoothing line generated by LOWESS (locally weighted least squares) locally weighted regression witha bandwidth of 10.


believe that our sample is large enough that a generalpropensity towards ketamine-associated respiratorydepression is unlikely.

The study is also limited in that we used a singleketamine administration protocol, i.e., 1.5 mg/kg IVover 1 minute. It is possible that hypoventilationmight be detectable with faster administration, andthis should be the subject of future study. We also didnot study IM administration, although given theslower absorption of ketamine via this route, hypo-ventilation would reasonably be even less likely.

CONCLUSIONS

No hypoventilation was observed in 20 ED pediatricpatients receiving ketamine 1.5 mg/kg IVadministeredover 1 minute. We found no evidence of respiratorydepressant properties for this dissociative agent.

References

1. Krauss B, Green SM. Sedation and analgesia for procedures inchildren. N Engl J Med. 2000; 342:938–45.

2. Green SM, Rothrock SG, Lynch EL, et al. Intramuscularketamine for pediatric sedation in the emergency department:safety profile with 1,022 cases. Ann Emerg Med. 1998; 31:688–97.

3. Green SM, Rothrock SG, Harris T, Hopkins A, Garrett W,Sherwin T. Intravenous ketamine for pediatric sedation in theemergency department: safety profile with 156 cases. AcadEmerg Med. 1998; 5:971–6.

4. Pena BMG, Krauss B. Adverse events of procedural sedationand analgesia in a pediatric emergency department. AnnEmerg Med. 1999; 34:483–90.

5. Green SM, Johnson NE. Ketamine sedation for pediatricprocedures: part 2, review and implications. Ann Emerg Med.1990; 19:1033–46.

6. Hamza J, Ecoffey C, Gross JB. Ventilatory response to CO2

following intravenous ketamine in children. Anesthesiology.1989; 70:422–5.

7. Tobias JD. End-tidal carbon dioxide monitoring duringsedation with a combination of midazolam and ketamine forchildren undergoing painful, invasive procedures. PediatrEmerg Care. 1999; 15:173–5.

8. Bourke DL, Malit LA, Smith TC. Respiratory interactions ofketamine and morphine. Anesthesiology. 1987; 66:153–6.

9. Maduska AL, Hajghassemali M. Arterial blood gases inmothers and infants during ketamine anesthesia for vaginaldelivery. Anesth Analg. 1978; 57:121–3.

10. Morel DR, Foster A, Gemperle M. Noninvasive evaluation ofbreathing pattern and thoraco-abdominal motion following theinfusion of ketamine or droperidol in humans. Anesthesiology.1986; 65:392–8.

11. Shulman D, Bar-Yishay E, Godfrey S. Respiratory mechanicsand intrinsic PEEP during ketamine and halothane anesthesiain young children. Anesth Analg. 1988; 67:656–62.

12. Soliman MG, Brinale GF, Kuster G. Response to hypercapniaunder ketamine anesthesia. Can Anaesth Soc J. 1975;22:486–94.

13. Virtue RW, Alanis JM, Mori M, et al. An anesthetic agent:2-orthochlorophenyl, 2-methylamino cyclohexanone HCl(CI-581). Anesthesiology. 1967; 28:823–33.

14. Zsigmond EK, Matsuki A, Kothary SP, et al. Arterial hypoxemiacaused by intravenous ketamine. Anesth Analg. 1976; 55:311–4.

15. Pandit SK, Kothary SP, Kumar SM. Low dose intravenousinfusion technique with ketamine. Anesthesia. 1980; 35:669–75.

16. Stanley V, Hunt J, Willis KW, et al. Cardiovascular andrespiratory function with CI-581. Anesth Analg. 1968; 47:760–8.

17. Smith JA, Santer LJ. Respiratory arrest following intramuscularketamine injection in a 4-year-old child. Ann Emerg Med. 1993;22:613–5.

18. Mitchell RK, Koury SI, Stone CK. Respiratory arrest afterintramuscular ketamine in a 2-year-old child. Am J EmergMed. 1996; 14:580–1.

19. Colman Y, Krauss B. Microstream capnography technology:a new approach to an old problem. J Clin Monit Comput. 1999;15:403–9.

20. Hagerty JJ, Kleinman ME, Lyons AC, Zurakowski D, Krauss B.Accuracy of a new low-flow sidestream capnography technol-ogy in newborns. J Perinatol. 2002; 22:219–25.

21. Sherwin TS, Green SM, Khan A, Chapman DS, Dannenberg B.Does adjunctive midazolam reduce recovery agitation afterketamine sedation for pediatric procedures? A randomized,double-blind, placebo-controlled trial. Ann Emerg Med. 2000;35:239–44.

22. Miner JR, Heegaard W, Plummer D. End-tidal carbon dioxidemonitoring during procedural sedation. Acad Emerg Med.2002; 9:275–80.

23. McQuillen KK, Steele DW. Capnography during sedation/analgesia in the pediatric emergency department. PediatrEmerg Care. 2000; 16:401–4.


ventilatory response during dissociative sedation in children—a pilot study

Documents